On Wed, Feb 24, 2021 at 4:26 PM Mike Kravetz <mike.kravetz@xxxxxxxxxx> wrote: > > On 2/18/21 4:48 PM, Axel Rasmussen wrote: > <snip> > > @@ -401,8 +398,10 @@ vm_fault_t handle_userfault(struct vm_fault *vmf, unsigned long reason) > > > > BUG_ON(ctx->mm != mm); > > > > - VM_BUG_ON(reason & ~(VM_UFFD_MISSING|VM_UFFD_WP)); > > - VM_BUG_ON(!(reason & VM_UFFD_MISSING) ^ !!(reason & VM_UFFD_WP)); > > + /* Any unrecognized flag is a bug. */ > > + VM_BUG_ON(reason & ~__VM_UFFD_FLAGS); > > + /* 0 or > 1 flags set is a bug; we expect exactly 1. */ > > + VM_BUG_ON(!reason || !!(reason & (reason - 1))); > > I may be confused, but that seems to be checking for a flag value of 1 > as opposed to one flag being set? (Assuming I implemented it correctly!) It's the logical negation of this trick: https://graphics.stanford.edu/~seander/bithacks.html#DetermineIfPowerOf2 So, it's "VM_BUG_ON(reason is *not* a power of 2)". Maybe the double negation makes it overly confusing? It ought to be equivalent if we remove it and just say: VM_BUG_ON(!reason || (reason & (reason - 1))); > > > > > if (ctx->features & UFFD_FEATURE_SIGBUS) > > goto out; > <snip> > > diff --git a/mm/hugetlb.c b/mm/hugetlb.c > > index 3bfba75f6cbd..0388107da4b1 100644 > > --- a/mm/hugetlb.c > > +++ b/mm/hugetlb.c > > @@ -4352,6 +4352,38 @@ static vm_fault_t hugetlb_no_page(struct mm_struct *mm, > > VM_FAULT_SET_HINDEX(hstate_index(h)); > > goto backout_unlocked; > > } > > + > > + /* Check for page in userfault range. */ > > + if (userfaultfd_minor(vma)) { > > + u32 hash; > > + struct vm_fault vmf = { > > + .vma = vma, > > + .address = haddr, > > + .flags = flags, > > + /* > > + * Hard to debug if it ends up being used by a > > + * callee that assumes something about the > > + * other uninitialized fields... same as in > > + * memory.c > > + */ > > + }; > > + > > + unlock_page(page); > > + > > + /* > > + * hugetlb_fault_mutex and i_mmap_rwsem must be dropped > > + * before handling userfault. Reacquire after handling > > + * fault to make calling code simpler. > > + */ > > + > > + hash = hugetlb_fault_mutex_hash(mapping, idx); > > + mutex_unlock(&hugetlb_fault_mutex_table[hash]); > > + i_mmap_unlock_read(mapping); > > + ret = handle_userfault(&vmf, VM_UFFD_MINOR); > > + i_mmap_lock_read(mapping); > > + mutex_lock(&hugetlb_fault_mutex_table[hash]); > > + goto out; > > + } > > } > > > > /* > > > > I'm good with the hugetlb.c changes. Since this in nearly identical to > the other handle_userfault() in this routine, it might be good to create > a common wrapper. But, that is not required. Makes sense, I can send a v9 with a helper for this defined. I'll wait until at least next week to do so, to pick up any other comments v8 may get in the meantime. Thanks! > -- > Mike Kravetz
